Give Me 33 Minutes, I'll Make You Disgustingly Good at Ad Creative
A lead-gen Meta ads operator walks through the full creative system her agency runs: where ideas come from, how every ad is broken into trackable building blocks, and how winners get found, scaled and iterated.
Posted
2 days ago
Duration
Format
Tutorial
educational
Views
669
30 likes
57 · 43
Big Idea
The argument in one line.
Winning lead-gen ads come from a repeatable system that breaks every ad into angle, audience, awareness level and execution style, so you know why winners won and can rebuild them on purpose instead of hoping AI finds one.
Who This Is For
Read if. Skip if.
READ IF YOU ARE…
You run Meta ads for coaches, agencies or service businesses where the goal is qualified leads and booked calls, not checkout sales.
Your account has a couple of winning ads but nobody on the team can explain why they won or what to make next.
You manage a creative team or freelancers and need a shared vocabulary for briefing, tagging and reviewing ads.
You've started generating ad scripts with AI and the leads coming in look worse than before.
SKIP IF…
You want a script-writing tutorial; the body copy and frameworks are named but deliberately not taught here.
You sell low-ticket e-commerce products with fast purchase feedback; much of the tracking advice is built for long lead-to-sale cycles.
TL;DR
The full version, fast.
On Meta today the creative is the targeting, so a lead-gen account lives or dies on how it produces creative. The system starts with research from lead forms, sales call transcripts, surveys, testimonials and long-running competitor ads, then tags every ad by angle, audience, awareness level and execution style in a simple spreadsheet. New angles get cheap $200 to $500 signal tests using stills. Ads should blend native and direct response, vary execution styles and messenger characters (mirror, aspiration, authority, character), and open with hooks built from a character, a prop and a scene. Winners that hit agreed KPIs move to a scaling campaign, and the tagged building blocks tell you exactly how to iterate on them.
Free for members
Chat with this breakdown — free.
Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.
AI is blamed for soulless ads and bad leads; credentials (10+ years, Linktree, Wayflyer, coaching brands) and the promise of a full system.
00:55 – 02:05
02 · The problem this system solves
Andromeda needs creative diversity, accounts recycle the same hooks, winners fatigue, nobody knows why they won, CPL climbs and sales complains about lead quality.
02:05 – 05:36
03 · Where to find winning ideas
Internal pools (winning ads, lead forms, sales calls, NPS surveys, interviews, testimonials) and external pools (competitor ads via Atria, Reddit, Amazon reviews, news, holidays). Sort competitors by longest running but don't over-index.
05:36 – 06:02
04 · Angle ideation
Every ad starts as a concept assigned to an angle category, which helps both planning and later pattern-spotting.
06:02 – 07:09
05 · The levels of awareness
Problem, solution and product aware, explained with a cough: each stage needs completely different messaging.
07:09 – 08:42
06 · The tracking spreadsheet
One row per ad: concept, angle, awareness level, sub-niche, campaign type, execution style, idea source and hypothesis. Keep it simple or the team won't use it.
08:42 – 13:45
07 · Angle testing
Zero-to-one accounts need cheap signal tests ($200 to $500) with messaging stills split by awareness level and sub-niche; one-to-hundred accounts mostly iterate instead.
13:45 – 16:35
08 · Ad execution styles
Four building blocks (angle, audience, awareness, execution). Blend native and direct response, vary formats so Meta doesn't lump ads together, start with founder face-to-camera or testimonial montage.
16:35 – 21:07
09 · Character and messenger
One-to-many communication theory: sender, channel, receiver, noise. The four messengers are the mirror, the aspiration, the authority and the character. Marketing equals psychology plus economics times technology.
21:07 – 24:15
10 · Hooks that stop the scroll
Hooks have a visual and a scripted half and run on contrast and tension. Break each into character, costume, prop and scene, like $100 frozen in an ice cube.
24:15 – 26:37
11 · Brand response scripting
Body copy blends brand voice with direct response, chooses a framework deliberately (PAS, AIDA, BAB and others) and wraps the message in a real market story. Win rate runs 10 to 20 percent.
26:37 – 28:29
12 · Killing losers and scaling winners
Agreed KPIs decide winners in a testing campaign; winners graduate to a scaling campaign like an all-star roster, and testing share shrinks over time.
28:29 – 30:13
13 · Lead gen vs e-com tracking
Lead gen has slow lead-to-sale feedback, so build loops from form-field combinations and sales-team feedback and judge on CAC, CTR, CPM and lead scoring.
30:13 – 33:25
14 · Iterating on winners
Recombine tagged building blocks: winning still to green-screen UGC to animated video, winning hook onto proven body, swap the character, convert to carousel. Outcome: a steady creative pipeline and lower CAC.
Atomic Insights
Lines worth screenshotting.
On Meta's Andromeda system the creative is your targeting, so creative diversity is how you reach new pockets of buyers.
Sales call transcripts and lead form answers are the best ad research because they capture the market's language before customers drink the Kool-Aid.
When studying competitor ads, sort by longest running, but don't over-index on them because most lead-gen advertisers have broken tracking.
If ten ads all look like the same founder face-to-camera, Meta treats them as one idea and they compete against each other.
A creative tracking spreadsheet only works if it's simple enough that the team actually fills it in.
Angle tests are data signals, not winners: spend $200 to $500 on stills with the message on them and read click-through rate.
The best ads right now look native enough to slip past the viewer's mental ad blocker and carry enough direct response to get the click.
Every ad has one of four messengers: the mirror, the aspiration, the authority or a character, and each one lands the same message differently.
Using your own generation's slang on an older market makes you sound like an imposter, and the audience filters you out instantly.
Marketing is psychology plus economics, multiplied by technology, which means AI only accelerates whatever psychology you already have.
A strong scroll-stopping hook can be as simple as a character the market recognises, an odd prop and a familiar scene.
A healthy lead-gen account hits a 10 to 20 percent creative win rate, and most advertisers aren't even tracking theirs.
As an account matures, the share of budget spent on testing should shrink because more of it flows to proven winners.
Hook rate and hold rate are indicators; CAC, CPM, click-through rate and lead scoring decide whether a lead-gen ad actually won.
A winning hook from one ad can be stitched onto a proven body from another to make a new ad with two tested halves.
Takeaway
Tag every ad so winners can be rebuilt
WHAT TO LEARN
Lead-gen creative stops being luck once every ad is broken into angle, audience, awareness level and execution style, tested cheaply, and iterated from the parts that proved themselves.
02The problem this system solves
Creative diversity is how Meta finds new buyers, so an account recycling a handful of similar hooks will fatigue fast and see lead quality slide.
Without knowing why an ad won, teams make creative reactively when results drop, and CPL and CAC climb as a direct result.
03Where to find winning ideas
Mine sales call transcripts, lead form answers and quarterly surveys first, because they hold the market's own words before customers absorb your messaging.
When researching competitors, filter by longest running ads but treat them as a weak signal, since most lead-gen advertisers have broken data setups.
04Angle ideation
Assign each concept an angle category up front so patterns across winners become visible later instead of staying buried in ad names.
05The levels of awareness
Problem-aware, solution-aware and product-aware prospects need very different messages, so label each ad with the stage it targets.
06The tracking spreadsheet
Track concept, angle, awareness level, sub-niche, campaign type, execution style, idea source and hypothesis per ad, and nothing more until the team uses it.
Treat every ad as an experiment with a stated hypothesis so a loser still teaches you something about the market.
07Angle testing
In accounts with no winners yet, run $200 to $500 angle tests using simple stills and judge them on click-through rate as data signals.
Test separate sub-niches inside one market, because each prospect needs to see themselves mirrored in the person on screen.
08Ad execution styles
The strongest ads look native to the feed so they slip past the viewer's mental ad filter, then carry enough direct response to earn the click.
Vary execution styles across the account; near-identical founder videos get treated as one idea and cannibalise each other's delivery.
Start with founder face-to-camera or testimonial montage, then add statics, carousels, native news ads and whiteboards alongside them.
09Character and messenger
Choose deliberately between the mirror, the aspiration, the authority and a character as the messenger, because each changes how the same message is received.
Speak in the market's own language; borrowed slang from another age group or class makes the ad read as an imposter.
AI accelerates the psychology and economics you already understand, so it multiplies bad thinking just as fast as good thinking.
10Hooks that stop the scroll
Plan the first three seconds visually before writing anything else, because a hook without a deliberate visual is the most common reason ads flop.
Build visual hooks from a recognisable character, a costume, an out-of-place prop and a familiar scene to create tension and an open loop.
11Brand response scripting
Pick a copywriting framework deliberately for each body instead of letting AI produce generic scripts that nobody wants to watch.
Wrap most scripts in real stories from the market, which is what pushes a well-run account toward a 10 to 20 percent win rate.
12Killing losers and scaling winners
Agree on winner KPIs before launching, usually a set number of leads or sales at a given spend, and kill anything that misses regardless of effort.
Move proven winners into a separate scaling campaign so, over time, a shrinking share of budget goes to testing.
13Lead gen vs e-com tracking
Because lead gen gets slow sales feedback, judge ads on CAC, CPM, click-through rate and lead scoring rather than hook or hold rate alone.
Build extra feedback loops from form-field combinations and sales-team input, remembering the sales team can have an off week too.
14Iterating on winners
Iterate on winners by changing one building block at a time: same angle in new formats, same script with a new messenger, still to carousel.
Combine a proven hook from one ad with a proven body from another to make a new ad where both halves are already tested.
Glossary
Terms worth knowing.
Andromeda
Meta's current ad delivery and retrieval system, which relies heavily on the variety of creative in an account to decide which people to show ads to.
Levels of awareness
A model from Eugene Schwartz's Breakthrough Advertising that sorts prospects by how much they know: aware of the problem, aware of possible solutions, or aware of specific products.
Angle
The specific way an offer is framed for a specific person in a specific situation, usually through a particular pain, moment or belief.
Sub-niche
A narrower slice of an audience, such as ceramic coating shops versus general mechanics inside the broader auto shop market.
Execution style
The format an ad concept is produced in, such as founder face-to-camera, testimonial montage, B-roll with voiceover, street interview, static, carousel or whiteboard.
Native ad
An ad designed to look like the organic content people already consume on the platform, so it doesn't trigger the instinct to skip ads.
Direct response
Advertising built to make the viewer take a specific measurable action now, such as clicking or filling out a form.
Brand response
A copywriting style that blends a company's authentic tone of voice with direct response structure.
Dog whistle
Language and references a specific market recognises as its own, which signal that the advertiser genuinely belongs to or understands that group.
CPL
Cost per lead: ad spend divided by the number of leads generated.
CAC
Customer acquisition cost: total spend required to win one paying customer.
Hook rate and hold rate
Video metrics showing the share of viewers who watch past the first seconds (hook) and how many keep watching further in (hold).
Lead-to-sales time
How long it typically takes for a new lead to become a paying customer, which controls how quickly an ad's real performance can be judged.
“We all know now that the creative is your targeting.”
one-line thesis, no setup needed→ TikTok hook↗ Tweet quote
13:10
“Somebody's on their phone mindlessly scrolling on their little dopamine rectangle, probably sitting on the toilet. They literally have an ad blocker in their brain.”
See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.
17px
metaphoranalogy
AI is sucking the soul out of your ads and it is attracting you shitty leads and clients. If you want to get disgustingly good at ad creative before you throw any AI into automations or speeding you up, first you need a proven system to come up with winners and actually scale them. That's what I'm going to be teaching you now.
I'm going to show you my ad creative system. I have been in marketing for over 10 years. I work with billion dollar brands like Linktree and Wayflyer.
I work with seven and eight figure mainly coaching brands. And what I'm very good at is getting highly qualified leads off of meta ads. Today, I'm going to walk you through the high level of my system.
Now, why this is important is because a lot of people are really good at e -commerce ads, but they haven't figured out how to translate that into lead generation. And ultimately, the sale starts in the click, finding that person.
We all know now that the creative is your targeting. So let's get into it. So we're going to be going through every single component in this system today.
As the kids call it, we're going to speed run it. But let's understand the problem we're trying to solve because all this is, it's an ad creative system. The system needs a goal, which is to find and produce winning creative.
Now, the keyword there is winning, right? The goal is not to just produce creative.
Instead, this system exists to solve a problem. So right now, Andromeda needs more creative diversity to find the right buyers. But the accounts that we work with are still relying on the same few hooks.
angles or formats. The team doesn't have a system for finding new buyer -led ideas. The creative gets made reactively when performance drops.
Meta has just a few buyer signals to work with. Winners fatigue fast. Nobody knows why they won.
The CPL and the CAC rise. The lead quality declines. Sales team complain about bad leads.
More calls get wasted on people who are never likely to buy and spend becomes harder to scale efficiently. So we work on anywhere from budgets on 10K up to like 600K.
And it all starts with the ad creative. Now with our system, this is just inputs, outputs, and outcomes. And so the very first thing that needs to happen is we need a good idea.
Now, the number one place to get your best idea is your current winning ads in the account. We're going to go over that soon and how to use those building blocks and how to iterate. For now, if you have a newer account, well, how do you come up with those new ideas?
Or when you need new ideas into your account, where do you get them from? So for us, we have a pretty pimped out AI system that goes through all of this data and then we still use our human brains. But I want to show you where are the pools of where you can look from.
So there's the lead form data, right? Like the actual language of the market. Your market are going to be your best marketers, right?
Your customers are your best marketers. Sales call transcripts, absolutely gold for finding new angles. And it's good to get those two because it's before.
your customers drink too much Kool -Aid. It's not a bad thing, but before too much. We also survey like NPS, like an in -depth qualitative NPS survey each quarter.
We interview our clients. So we're getting that gold out of their brain. We'll look at top organic content.
We're getting all of the testimonial video transcripts in, and then there's the winning ads. Now we're also looking at external data. So this might be looking at other client winning ad swipes, whatever, Reddit forums, Amazon books.
What's the top performing book? What are the comments on there? What's happening in the news and media?
And I think what a lot of people miss in Legion is still like. what holidays, like what trend can you ride? And generally, what is the zeitgeist?
So when we're looking at other competitor ads, yes, that's good. But all you're really doing is like recycling someone else's idea, right? But when we do look at competitor ads, we use Atria.
So I love Atria. We use it to like actually see all the way down to a buyer on our ads, qualified leads, everything, not just the leads.
And you can use this. And let's just say that we wanted to look up coaching. It's going to show me all of these different coaching ads.
And I can also do it for how long it's been running. So it can be most relevant. or it can be the longest running.
I would always use longest running and obviously you can get more specific, et cetera, as well. So what I want to do is find, well, if someone's working either directly or indirectly as a competitor, I want to see if I can swipe that. You can also do it for free in the Facebook ads library, of course, but I prefer Atria or I think there's like another tool like Foreplay and stuff as well.
So we look for the longest running, especially in lead generation. over index competitors. I think in e -commerce, there's a lot of really sophisticated players in ad creative.
I don't think that's the same in lead gen. I think most people don't know what they're doing. And even sometimes they have a long running ad and they don't have their data set up.
I know because I'm auditing all the accounts, like I'm auditing all of these accounts and restructuring the data infrastructure because they've been doing it wrong. Now, the next thing is, yes, it's important to have these ideas. And my best team members, that are really good ad creative strategists are just into some weird shit.
And they watch a bunch of different stuff. Like our best ones, they're like studying a movie every night. You know, they're like reading weird and different types of material and they're doing what AI can't do yet.
They're like connecting this creativity and these synapses in their brain. So that's the research. Inputs, outputs, outcomes.
If we're not putting good information, good data, good research into the system, nothing else matters. So once we've had that initial research, well, the next is that. angle ideation.
So for us, when we are creating all of our ads, we want to be able to see the core components. I'm going to summarize for this, this for you soon, and you'll see it. I'm slowly revealing it.
But when we're choosing our angles, we have the ad concept, and then we have these different angle categories. It helps us to plan, but it also helps us to categorize. The next thing we're doing is we're thinking about, well, what level of awareness is this for?
And this is from, I'll zoom out for a second. This is from Eugene. what's this breakthrough advertising, right?
When we do our ads, the creative is the targeting and we want to be able to mix and match these different things. And we want to be able to see the patterns of what is winning. But what might happen is we have a problem aware ad.
So it's like a problem. Problem aware is like, I've got a cough. That's it.
I know I got a cough. Next is solution aware. Okay.
I've got a cough. I can go to a doctor. I can get cough drops.
I can get a... a honey tea, right? I'm assessing different options.
And then product aware is, okay, I've got a cough. I want to go to a doctor. I am assessing competitors.
I can go to Dr. X, Dr. Y, Dr.
Z. And your market is in these different levels of awareness, these different stages of awareness and how you can imagine, right? How you speak to each of them is going to be very, very different.
And you don't want to just make your ad. Without this strategy, marketing is an art and a science and ad creative is a really prime example of that, of blending that art and that science because you want to be very scientific about your testing. So one of the ways that we do that is just by having a simple spreadsheet.
that has our ad concept, that has our angle, our awareness level, the sub -niche or the audience, you'll hear me calling it, the campaign type and the execution style. There's a few other things that we track, but why do we do this? So I audit big accounts that spend lots of money and they don't track this.
So what happens is, is their account usually scales and there's a couple of winners, but nobody really knows why it wins and they're never iterating. on those winners, which we're going to talk about more soon. And they're not iterating on those winners because they don't know why they won.
They accidentally found a winner. And when you can't replicate success, it's very hard to scale. So when you're installing the system, The main thing is, I will say, is to keep it simple, especially this spreadsheet.
I made like a crazy spreadsheet with like hook and hold rates at the seven day and the 21 day and all of this stuff. And when I first gave it to my team, it was like a meltdown. It was too hard, right?
Too much change management. The other thing is every single ad is an opportunity to learn. Everything we do in marketing is an opportunity to learn.
There's no failures. There should only be learnings. And so everything in marketing, we should have like a hypothesis.
Like, where did you come up with that? So one of the other things we test is like, where was the idea from? Because then we can trace good idea pools.
But it's like, okay, what's the hypothesis of this ad? And if it won, why? And if it didn't, why?
What were we testing and why? So that's the ad angle research. That's the ideation.
And now I want to teach you something around. angle testing, because there's two different types of accounts, right? There's basically a zero to one account.
So sometimes we're working with a client that is a large company, but they've never cracked paid social. Maybe they've worked with two, three, five agencies, an internal team. No one's really ever cracked it, right?
They got referral or organic, whatever, cold calling, and that's working. So if you're starting from zero, you don't have... winning ads to dissect and iterate on to start.
You have to get some sort of data that's going to get you moving. Now, when it's one to a hundred, really testing ad creative, you only like will do this angle testing if things are not becoming winners. Otherwise everything else should be iterations and coming up with the new angles over there.
So how we measure success of this campaign of this angle testing is that we'll blend click -through rate. and quality of leads pending volume. But usually we're looking at click -through rate and like early signals.
We're only spending like $200, $500 on these campaigns just to get signals, data signals. These are not winning ads. These are data signals.
Now, this is where you're going to start to see these concepts, these building blocks of the ads. And we're going to be going really deep on these. So when we're doing our angle testing, typically we're just doing something like this.
I learned this in a Ryan Dice mastermind many, many years ago. And it's just to have a simple still with your messaging idea on here, usually broken down.
We break it down into levels of awareness and then that's it. So you can either do these like selfie face to camera and put them on this. It can sometimes depend on the client that you have, but you have to be purposeful about what's going into the ads.
So for example, there's the sub niche or the audience. You can see in this example, this is, I think it's called like a cart, like car wrecker.
Like they fix like cars that have been in accidents. This guy does like the wraps, car wraps. And this guy is like more of a run of the mill mechanic.
Three totally different sub niches. The sub niches are important. The audience is because they're a mirror, which we're going to talk about soon.
They're a mirror and someone's going to see themselves in them. And if you're only testing this bloke, this guy, and you're not testing these, you're potentially losing. money, right?
You're not finding these pockets of audiences just in general. We'll talk more about that soon. Then we have the angle testing.
We have the offer testing and then add execution style that gives you a little taster into the building blocks. Okay. So now you have like your thought process.
This is what we're doing. This is what we're testing. Now you should have like, basically you have this kind of like mapped out, right?
And now you're getting ready to start. creating the ad. So you have to understand, well, what are the core building blocks of the ad?
So to summarize some of the things that we've just spoken about, that is your angle, that is your audience, the ad execution style, and the awareness level as well. So when we think of our ad, it's an ad concept, right? It's a concept.
And then Then we want to break down that concept into its core component parts so that we can reverse engineer why it did or didn't work. So we have the audience and the awareness level.
So we want to be able to find profitable sub -niches and we want to be able to understand what level of awareness that is in. For example, auto shop is like an audience, but the ceramic coat. versus a mechanic is a sub -niche.
And this is like, I'm teaching you our internal jargon. Every company and every marketer is a little bit different. The angle then is the specific way we present the offer to a specific person in a specific situation.
It's like the entry point into the market's problem. And it's going to frame your offer through a particular pain, moment, or belief. And we test the idea across multiple formats, right?
So we've got this, we've got the angle, and now we're going to have multiple formats. So there's each ad execution style. I'm obviously speaking about videos here.
Each ad execution will have a hook, a body and a CTA. Overall, with our ads, what we are really seeing working is having this blend of native and direct response. Native and direct response.
Why? We need to zoom out. So contextually, you always have to think about your ad in context, right?
And so contextually, what's happening is somebody's on their phone. mindlessly scrolling on their little dopamine rectangle, right? Probably sitting on the toilet and they don't want to look at ads.
They literally have an ad blocker in their brain to stop seeing your ads. They don't want to watch anything that stinks like an ad. They're not interested.
So we're going to have to like honey trap them a little bit, right? We have to think, yeah, we're buying media. on a native platform.
And that ad needs to look like something that they would want to consume. And it has to have a direct response flavor to get them to take action. So those are the best ads working right now.
Now let's dive in to some of these. We've got the execution style. So we've got videos, right?
Now the execution style is the way that we... cook up the video and a lot of people stop here they're like yeah found a face to camera but they don't understand the other building blocks what we also want to make sure that we do in the account is have a blend of different execution styles so i'm going to be talking about iterations and how to do that soon but effectively if you have all found a face to camera and they all look exactly the same andromeda is not going to go out and test them against everybody else in the market is just going to go, all of these ads look the same.
They all compete against each other and nothing works. So when we're building out our ads, we even do. and I'll talk about this in a second, very different visual hooks, very different scripting of hooks for each angle, for each body in our ad execution style.
Now, also what you want to be able to do is you want to be able to identify on your account what type of creative is going to work. So typically, if you need somewhere to start, founder face to camera, if they've got a bit of charisma about them. testimonial montage.
Those are going to be the best. If you don't have those assets, I would do AI ambassador tells a story or B -roll and voiceover would be the other thing. We also really enjoy like these like street interviews as well.
Now, of course, you should also have statics in your account and have a mixture of statics. For example, something like this as a carousel is crushing for us on retargeting. These sorts of native news ads crush for us on pretty much I would say 90 % of accounts that we run, this ad will work.
Whiteboards, you should be testing a bunch of different ads. Now, what you'll want to do is mainly be building out the ads that work, do more of what works. And this is such a thing that I see because when you do ad creative, it's kind of in the title, right?
Like you want to be creative, but what you actually need to do is be scientific ads. ad scientific. How would you call that?
You want to be an ad scientist. And once you find something that's working in your formula, you want to double down on it. So what we tend to do is depending on the account and how many winners we have in the account, I'm going to talk about this soon.
We just want to keep redoing, redoing what's working. However, we want to also find what we can test and what we can add into the account.
So as an example, we might test claymation. We might test founder AI. We might test a reaction video, whatever, like these different formats to try and find them on the account.
Now, the next layer deep on that, hopefully I'm not getting too, too nerdy for you guys. I know it's a lot of information, but I want you to make good ads is the character and the messenger. Okay.
Look, cause What you have to understand, again, is like context. Whenever you put a piece of marketing out there, you're doing communication.
But instead of doing it one to one, you're doing it one to many. And you have to understand the nuances of one to one communication versus the nuances of one to many communication and what that actually means. So in one to one communication, you have you, the sender, right?
And you are going to put out a message and it's going to get filtered through the channel or the media. So if I verbally tell my husband something like, hey, you shouldn't have done that.
That wasn't really good, whatever. But if I message that, that media, that channel can come across totally different, which is why sometimes phone calls are better, right? Then you have the person decoding that message.
So that's the receiver. And so what happens is like, have you ever told your mom or your partner or like somebody in your team, something, a client, and they've reacted really badly to it? And you're like, that's so weird.
How, like, how have you taken my message and like, not got what I'm telling you? And that's because you have to remember the receiver, they're going to receive your message through their cultural lens, through their mood on the day, all of these different things. And so oftentimes what I see, I might see like, Gen Z or whatever market is.
And they'll be talking to a more mature audience, like older, but they'll be using words like speed run. They're using like their own lingo, right? So when we're doing this, we have to be using, it's what's called the dog whistle.
of that market. Because when we create ads, we don't want them to come across as an imposter. People can sniff it out straight away.
It's like if you're talking to rich people and you're using a certain language, they're going to filter it out because they're going to think you're an imposter. So going back to this, they give a response, they give you feedback, and that should make you the sender better at giving messages. You have to also remember that whenever you're putting a message out there, there's a bunch of noise, politics, other ads.
They're hungry. They got bills, right? They're living a life.
You're literally demanding their attention amongst all of this noise. So all of that to say, the sender is very important. The sender is very important.
And when you want to get a message to your market, there's four different senders you can have. And again, I've never really seen anybody break it down like this, but I love systems and I love frameworks for thinking. I don't know if that's because I'm lazy and I don't want to have to reinvent the wheel, but I see a lot of people never building a system.
So they never can speed up. They can never replicate that success. And you can kind of imagine that when you start to throw AI into this, but you don't have these levels of frameworks and thinking, how is AI meant to act?
AI is just... your team member. And like, how would you train a human?
Well, you've got to tell them how to think and you've got to give them examples of success. If you can't do that, the AI no worky, right? You're going to get slopped.
So back to this, there's four different characters you can have in your ads. There's the mirror. So for example, if I was working on an offer for breastfeeding mums, The mirror would be a mum that's currently breastfeeding.
The aspiration might be, but in pain. The aspiration would be a mum, maybe the baby's like no longer breastfeeding, but it's a really healthy baby. And she talks about her like success journey.
The authority would be a doctor. And the character is, I might do like one of those baby podcasts, you know, when like the babies are talking on a podcast. I think they're a bit weird.
I've never used those, but as an example, those are the four different examples. And each one of them, the market, the receiver will interpret their message differently. And again, it's like marketing is not tasks.
Marketing is not running ads. Marketing is not just building a funnel and the tech. Marketing is a blend of psychology plus economics times.
technology. So to go back to the thesis of this video, AI is probably killing your account, it's because AI is technology. AI is an accelerator of that core equation, psychology plus economics.
And so what I'm going really deep on right now is the psychology component of things, the art and the science of actually persuading humans at scale to take an action you want them to do. like, it's just like, not just a super easy, simple thing, right? Okay.
Let's get into hooks just to draw it back. Cause I know I've seen this a million times, but this is your first rodeo. Zooming out.
We have four building blocks in our ads, angle, audience, ad execution, awareness level. When we're talking about the ad execution, which is where we are now, you have your ad execution style, which we went through. You have your hooks, your body, and your CTA.
So we've just gone through the ad execution styles. Now we are talking about the hooks. And so I don't know if you guys know who Adli is.
She like, this is like amazing content creator. She's huge. And I saw her do a talk in Nashville a couple of years ago.
And she does like, like, I think she runs a company called Viralish. And she was like, if somebody brings me a content concept and they don't just tell me what is the visual and what is the like. Like I think it was like just the visual.
I just wonder what is the visual in the first three seconds. I will not listen to them because I know it will flop because they haven't put their amount of time and effort into the most important thing, the hook, the visual of that. So when we think about hooks, there's broadly speaking two components.
There's the visual component. There's the scripting of that as well. And when we create hooks, what we want to be able to have is this contrast.
and this tension because you're basically trying to like gamify somebody's brain to get attention, right? So one way that you might hear people talk about this is like a pattern interrupt because you need to like get their attention. One version of this is distorting a human face.
Like if there's something on somebody's face, our brain is naturally wired to look at that. But what else are we naturally wired to look at? Now, this can depend on your market and what's familiar to them.
And I remember Adley talking about, you know, in one of my videos, she was saying she put tampons in a fridge. That's weird. It's not meant to be there.
It creates tension, right? So when we're doing our hooks, again, because like we break down all of these core components. So, for example, in this hook, we had a character.
We have a costume. So it's someone in high vis because this was for. this was for like, I think it was like for like air conditioning installation or something.
We had a prop and that character and that prop create a scene, right? And then what happens is like that falls and breaks and something. Again, character, we got old mate in a costume.
Like it's like, yeah, he's got, he, he is the mirror. He looks like the market we're trying to talk to. And the prop is that there's a hundred dollars in an ice cube.
It opens intrigue. It opens a loop. It's a pattern interrupt.
It's all of these things. And it's in a scene that our market would be used to seeing. Now it can just be more basic like this.
Hoplond always converts. That's the character. The prop is like this waffle iron thing.
And this is Australian money in case anyone's wondering. And it's like the hook is like putting that inside this waffle iron and that's like the scene. And that's it.
Like it doesn't even have to be that. elaborate. It's just something weird.
Okay. So that's the hook. Now I'm not going to go too crazy into the body, but the writing style that we use is what we call brand response.
It's like a blend of an authentic tone of voice of the company with direct response. But what a lot of people get wrong is they kind of just give AI and they shit out ads.
That's the best way I can describe it. They just write the most unengaging, uninspired trash ads that no one wants to watch or listen to. They're just ticking the box.
And instead they need to be thinking about, well, what is, what is like the modular way of writing this script? Because I want to know problem agitate solve. Is that going to be, is that going to be the best thing?
for my market? Or is it ADA or BAB or PASTA or SSS or FAB? Like these are the different frameworks that you want to be able to use.
So there's that. I'm not going to get into script writing today. Now, why do our ads work?
Like us as a company and our ads have like a, I would say across accounts and depending how mature they are, like a 10 to 20 % win rate, which is a lot. Most people aren't even tracking win rates, right? Like they don't have this into a system.
They're just launching shit and hoping it will work. But the way, the reason ours work is because we wrap them in a story, story selling. We wrap things in a story.
And so a lot of these scripts, unless it's like a direct to offer angle, most of our scripts are wrapped in a story, right? And it's like. legitimate stories from the market that are going to take root in somebody's brain and get them to want to pay attention.
We've been telling stories to engage each other since we were like cavemen and it's still the same, right? And again, I know this is like super deep and all you guys typically like what people want to hear are like these silver bullets for making ads. That's not it.
I'm not here to sell you that. Like all these gurus that are telling people it's easy, use AI, just launch this stuff. Not if you don't know what you're doing, because the AI will put you into a psychosis.
I've seen so many businesses in these AI psychosis because the AI tells them it's good. It ain't good. Like it ain't good at all.
You need to learn this stuff if you want to seriously compete. That's my message there. All right.
Next, let's get into killing losers and scaling. Killing losers and scaling winners. So you're going to launch your ads.
If you're good, you'll have a 10 to 20%. hit rate where they're actually winners. So how do you actually know they're a winner and what do you scale?
Well you're going to put everything into a testing campaign and you're going to use agreed upon KPIs to determine what is a winning ad. Always you always have to have the same page of goals and vernacular. So for us a winning ad will typically hit x number of conversions which is usually leads, especially because a lot of our clients have a longer lead to sales time.
But if they have high volume, shorter lead to sales time, we would use sales with a Y ad spend. And if they don't hit that, you have to kill your darlings. Like it doesn't matter if you're like, oh, but I tried really hard.
Sometimes it is the lamest ad that you never thought would work. That is the one that scales, right? Marketing is a humbling sport.
welcome, welcome to the club. Now, once you have those winners that are hitting that, then you just move them into a scaling campaign. And so I think of this like your NBA team, right?
What we're trying to do is have like top performers. They're getting trialed and tested. They're going through their training camps.
They're coming in, they're playing games. We're seeing if they're any good and the best ones make it to the all -stars team. Everybody else, pack your bags, you're going home.
And all we're trying to do is get as many people as possible into these winners because then we know we can scale the account, right? And even if your goal isn't to scale an account, you're going to be murdering your ad spend unless you have ads that win, right? So over time, your percentage of testing, if you are scaling, even if you're not, it should get smaller and smaller because you are putting your budget into winners.
Now, let's just talk a little bit about tracking this because I just don't see much content. And that's why I'm making this video.
I just don't see much content about lead generation ads. I see a lot about e -com and they're really good. A lot of the e -com guys and gals are obviously really, really good, but it doesn't always translate into lead gen.
And that's why you can't copy them blindly. You need to know how to think. So with the e -com versus lead gen, what do we need to understand?
Number one is the lead to sales time. When you have a When you have a product that you're selling, right?
When you have a product that you're selling, you get very quick feedback. It's a $30 product and does it sell? Yes or no.
So we don't always get that feedback. Instead, we need to create different feedback loops in lead gen to tell us whether that ad is a winner. So that could be the form fields, like what combinations of form fields are most likely to produce a buyer.
And it can... layer that with feedback from the sales team as well. Because you have to remember the sales team can have a bad week.
They just don't close anything. The leads were qualified, but you guys didn't close. So there's more moving parts.
So we will look at CAC. So the cost to acquire a customer. We'll look at click -through rate, CPM and lead scoring.
Yes, hook rate and hold rate is okay, but it's not key to the performance. They're indicators. Awesome.
Okay, cool. So I'm just going to show you a quick screenshot of Atria where we literally have all of the ads. What do they spend?
How many leads? What are the costs per lead, the appointment scheduled, and then the purchases? Like we're always tracking down to ideally as far down as the funnel as you can.
In e -com, that's a very short loop. Did they buy? Yes or no.
And it happens usually pretty quickly, right? For us, it can be a little bit different. All right.
Once you then, you've launched these ads, you're starting to see winners, you're starting to kill your losers. Now you have to get into the part that... most people never do, which is good for me because a lot of big accounts come to us and we get to install this and there's heaps of low -lying fruit, which is awesome, but it's iterating on winners.
People don't know how to iterate because they don't know what I just showed you, which is what are the core components. They're not tracking that. So you don't know what to make next.
And so basically, how can I explain this? It's like you have all these Lego blocks, right? Angle, ad execution style, awareness level, audience, and then you can like Go, okay, in this ad, we had a winning angle.
Let's take that angle and let's try it in three more execution styles. Okay, on this ad, we had a winning hook. That was it.
The hook was a 56 % hold rate. But on this ad, we have a proven body that we've re -hooked five times. let's put them together and make a new ad, right?
So let me show you this a bit more tangibly. On this account, we had a winning still ad. This is like that native news I was showing you earlier.
Then we iterated on it by doing like a green screen UGC, like talking about it. And then we iterated it again by animating the video, like animating that image into a video with AI. Done.
Like that's it. So how can you actually iterate? I don't have time to go through our whole system.
So here's some ideas. Number one is you can turn a winning hook visual into a still. So if you have a winning hook, just take a screenshot and make it a still.
If you have a video where the ad execution is a talking head, you can then put it into a voiceover with B -roll. If you have a winning ad with a character that's the founder authority, try the same ad with a different character. Try it with an aspiration, which is kind of this same script change character.
If you have a static or a video, make it a carousel. If it's a video script, text on still. Your goal here is to...
I've got it down here. Flog your winners. I don't know if that's an Australian thing, but flog your winners and understand why they worked.
The whole game of any success is to replicate what's winning and do more of it. But a lot of people struggle with that because it's boring. Or I've had ad creative strategists who are super creative, like their brains are just brilliant and wonderful and weird, but they won't do this stuff.
So they don't last very long because you have to iterate on this. It's an art. And it's a science.
And then the outcome when we actually have this working, which should be your goal of this system, is to build a consistent creative pipeline to prevent the added fatigue, launch stronger creative faster, find more buyer producing winners, improve your lead quality, reduce wasted sales calls and improve utilization rate, and close more deals and lower CAC.
which means you can scale spend more efficiently and increase your ROI. So the success is we know how to make new angle ideas. We know how to do iterations and variations, and we have this system to hold it all together.
That is really the goal. Now, of course, as we all know. the ad just gets the click.
It's literally worthless unless you can convert that click into a qualified lead. So watch this video next and it's going to walk you through my sales process that I have painstakingly refined after like a thousand different funnels. See you on the next video.
The Hook
The bait, then the rug-pull.
The video opens by blaming AI for bad leads, then spends 33 minutes arguing the real problem is older: most lead-gen accounts have no system for making creative, so nobody knows why the occasional winner won. What follows is the agency's whole pipeline, from research to iteration, sketched on a whiteboard.
Frameworks
Named ideas worth stealing.
02:00model
Inputs, outputs, outcomes
Research inputs
Creative outputs
Business outcomes
The creative system is framed as a pipeline where poor research inputs make everything downstream irrelevant.
Steal forany content or ad production workflow
02:20list
Idea pools
Current winning ads
Lead form data
Sales call transcripts
Quarterly NPS surveys
Client interviews
Top organic content
Testimonial transcripts
Competitor ads
Reddit and Amazon reviews
News, holidays and zeitgeist
Internal sources capture the market's own language; external sources supply trends and proven competitor angles.
Steal forad and content research briefs
06:25model
Levels of awareness
Problem aware
Solution aware
Product aware
From Breakthrough Advertising: the further along the prospect, the more specific the message must be.
Steal forsegmenting ads and landing pages by prospect stage
14:05list
Four building blocks of an ad
Angle
Audience / sub-niche
Awareness level
Ad execution style
Tag every ad with these so winners can be reverse-engineered and recombined.
Steal forcreative tracking sheets
14:30list
Hook, body, CTA
Hook
Body
CTA
Every video execution splits into three swappable parts.
Steal formodular video ad scripting
19:30list
The four messengers
The mirror
The aspiration
The authority
The character
Who delivers the message changes how the receiver decodes it; test the same script across messengers.
Steal forcasting UGC and testimonial ads
20:35concept
Psychology plus economics, times technology
Technology including AI multiplies the underlying psychology and economics; it cannot replace them.
Steal forexplaining why AI tooling alone doesn't fix marketing
23:20list
Hook anatomy
Character
Costume
Prop
Scene
Visual hooks are assembled from a recognisable character, an out-of-place prop and a familiar scene that create tension.
Steal forbrainstorming pattern-interrupt openers
27:00model
Test then scale campaign structure
Testing campaign with agreed KPIs
Kill losers
Graduate winners to scaling campaign
Shrink testing share over time
Likened to an NBA roster: trials and training camp, then the all-star team.
Steal forstructuring any paid media account
31:20list
Iteration moves
Winning hook visual to still
Talking head to voiceover plus B-roll
Same script, different character
Static or video to carousel
Video script to text on still
Winning hook plus proven body
Ways to flog a winner by changing one building block at a time.
Steal forextending the life of any winning creative
CTA Breakdown
How they asked for the click.
VERBAL ASK
33:05next-video
“The ad just gets the click. It's literally worthless unless you can convert that click into a qualified lead. So watch this video next.”
Bridges the creative system to the sales process video, framing the next video as the missing half of the funnel. Description links an agency coaching program and a done-for-you agency.
Add Modern Creator as a preferred source and Google shows you more of our breakdowns in Search, Top Stories, and AI Overviews. It only changes what you see, and you can undo it in your Google settings anytime.
Add to Preferred SourcesOpens your Google source preferences with us pre-loaded. Tick the box and you're done.
Ultra's creative and growth lead explains how a nicotine-free pouch hit a $100M run rate in under a year: customer-language angles, a three-part replacement formula, creators as strategists, and constant new footage.
A paid-social agency opens a real Meta ads account on screen to show why three to five ads carry most accounts, and the weekly system that keeps new winners coming before the old ones fatigue.
A performance-agency owner runs his actual research-to-render ad pipeline live, from a nine-hook swipe file to a chat prompt that renders 3D kitchen scenes in Blender.
Jeremy Haynes breaks down what actually controls Meta ad targeting, why your pixel isn't as smart as you think it is, and the exact 'Thunderdome' testing system he uses to find the 1-3% of ads that truly scale.